Update from the workshop:

 

It’s well known that things move. Buses move people, you move the objects you interact with, and you move yourself around. Being in control of what, when and how things move is a superpower. Controlling the bus timetable allows you to improve, or diminish the productivity of Sydney… controlling how quickly you raise your coffee cup to your mouth can make the difference between a gentle wakeup, or a miserable morning cleaning your shirt… The list goes on. In manufacturing, the more you control, the better your results become. The easiest way to visualise that is that manufacturing parts is like having a chain. The more demanding the parts, the stronger the chain needs to be, and the chain is only as strong as the weakest link. Controlling each link in the chain is the first step to identifying the weakest link and improving the strength of the chain. Practically, this involves purchasing high quality equipment, software and measuring tools. It also extends into monitoring your workshop environment, your vibrations, your temperature and secondary things like humidity, cleanliness/dust control. Temperature is one part of that chain that is a deep deep rabbit hole. Just like a scheduler can move buses around, or how you can move your coffee cup, temperature moves everything. Shifts in temperature cause materials to grow or shrink. This is what engineers call the “coefficient of thermal expansion”, or CTE. It’s measured in microns per metre per degree. Most steels have a CTE of around 11 microns per metre per degree. This means that if you had a 1 metre long rod at 20 degrees Celsius, left it our in the sun so it was around 40 degrees Celsius, it would expand and be 220 microns longer! That’s 0.2mm, or about the thickness of two sheets of paper. For some more perspective the height of the Sydney Harbour Bridge can move more than 100mm on a hot day, as all the steel expands! Every material has it’s own CTE. Teflon, for example moves 100 or so microns per metre per degree, so if my rod was made from teflon, and I left it out in the sun, it would end up being 2 whole millimetres longer at 40 degrees C! This has always been a problem in horology. The moment of inertia of an oscillator, such as the pendulum of a clock, is strongly tied to the distance that the mass is from the pivot point. The moment of inertia in a clock directly dictates the time keeping accuracy. So, maintaining a stable moment of inertia, and therefore a stable pendulum length is critical. Once horologists knew this, they started the hunt to identify materials that had very low CTE’s. Bizarrely, wood has a fantastic CTE if you measure it’s expansion along it’s grain. About 3 microns per metre, per degree, or about 3x better than steel. Unfortunately, humidity changes have disastrous effects for wood’s expansion, so it was quickly ruled out for most applications. Glass was then explored, with fantastic environmental stability, and relatively good CTE (around 6um). Glass was improved on as a pendulum rod material by using fused silica, or quartz, which has a CTE of just 0.55um! This technical exploration was then transferred to watchmaking, where companies have been using very special materials, such as invar, silicon, and even diamond to improve the chronometric accuracy of their timepieces. In the summer of 2023 and 2024, NHW has a different problem with temperature… As Sydney warms up, our workshop goes through very large temperature shifts. We do a few things to combat this, insulation, air-conditioning, and the most extreme, limiting machine usage. The machines in our workshop all consume electricity, and generate heat. That heat is then extracted either passively by our air-conditioning, or actively by chilling units and refrigerators. Our Kern Pyramid Nano, the first milling machine we purchased, has an active temperature management system that cools the machine down to a stability of +-0.5 degrees. The newly purchased Kern Micro HD also has a similar system, but it’s 10x as powerful, cooling the machine down to a stability of +-0.05 degrees! Both of these systems consume a very large amount of energy, and even more energy when they have to fight against 35 degree days… It’s a vicious cycle. As the outside temperature increases, the machines need to use more energy to cool down, which generates more ambient heat, which our air-conditioning systems need to fight against, which uses more energy… The limit is our available current draw to our industrial property. At one point, our distribution board taps out and we can no longer effectively run the workshop. The only choice? To turn off the machines and do other tasks. Cleaning, organisation, manual work, decoration of components, etc. But there is a clincher… and it’s a particularly nasty one. Our newest milling machine, the aforementioned Kern Micro HD. The Micro HD is the most accurate 5 axis milling machine in the world, and we have the only one in all of Asia. It’s a fantastic achievement, but this machine was designed to be run in a very specific, very controlled way. This is to a drill press as what a chicken is to a velociraptor. One of main differences between this machine and a machine in a lower price bracket is it’s fundamental construction. The machine is made from a mixture of aluminium, granite and iron. The moving components, such as the X,Y and Z axes are made from aluminium, but those aluminium parts are actuated with powerful magnets made from iron (linear motors). This effectively creates a bimetallic strip. One side is iron, one side is aluminium. Iron has a CTE of around 11um, and Aluminium has a CTE of around 22um. The baseline temp for the machine is 20deg- When the machine is “on” it is being actively cooled down to 20 deg within 0.05deg. But when the machine is off, it normalises to whatever the room temperature is. This can cause some serious issues… If the ambient temperature rises to above 28 degrees while the machine is not being actively cooled (off) then the iron-aluminium construction of the machine bends to a point where it causes permanent damage to the machine frame. The ultra-technical explanation is that as the machine is warming up, the frame bends in a predictable way, but as it cools down, the friction forces between the iron and aluminium mean that it deforms in an unpredictable way. The bottom line? If the machine reaches 28degrees or higher, it needs to be completely recalibrated by the manufacturer. A cool 30 thousand dollars in airfares, travel time, and recalibration costs. To make a long story even longer, the obvious solution: keep the machine on all the time, doesn’t really work… It costs approximately 40 dollars in electricity and consumables per hour to keep the Micro HD on in a “idle” state. We work an average of about 50 hours per week, which leaves 118 hours per week of idle time, which is 4720 hours per year, which is about 250k a year running cost to keep the machine “idle”. Turning the machine off, doing manual work and allowing our poor little factory to breathe unburdened during a hot day is the lesser of two evils… In our precision chain, some things are easier to control than others. As we continue down this path of manufacturing in Australia and go deeper down the rabbit holes, we are slowly realising that the things you thought were the least conspicuous (temperature!) cause the biggest problems. If you’ve made it this far, congratulations. I don’t expect all of you to listen to technical rambles like this, but if you have, I hope you appreciate another little peak behind the curtain of what it takes to make watches in Australia. All this makes your decision easy… why would you buy a watch made anywhere else? Josh